MySQL - Select Field With Nested 'WHERE' - mysql

Question

I'm working on a project which uses the SUM command to get a number of values. Now, all of this works fine but there is an issue when it comes to load time as the query takes 3.4 seconds to complete.
Here is an example of what I have so far:
SELECT
p.`player_id`,
p.`player_name` AS `name`,
d.`player_debut` AS `debut`,
SUM(a.`player_order` <= '11' OR a.`player_sub` != '0') AS `apps`,
SUM(a.`player_order` <= '11') AS `starts`,
SUM(a.`player_goals`) AS `goals`
FROM
`table1` r,
`table2` a,
`table3` p
LEFT JOIN `table4` d ON p.`player_id` = d.`player_id`
WHERE
r.`match_id` = a.`match_id` AND
a.`player_id` = p.`player_id` AND
r.`void` = '0'
GROUP BY
a.`player_id`
ORDER BY
p.`player_name` ASC
Cast your mind to line 4. That field is retrieved by making use of the LEFT JOIN further down the query. By taking those two lines out, load time decreases to less than 0.5 seconds - a significant improvement.
What I'm trying to achieve there (line 4), without success, is something similar to lines 5-7, where a sort of invisible WHERE clause has been applied.
The idea would be t4.date WHERE t2.order <= '14', but I'm not sure how I'd be able to get this to work without the aforementioned LEFT JOIN and increased load time that comes with it.
For clarification, here is how table4 was created - with the following query turned into a VIEW.
SELECT a.`player_id`, m.`date` AS `player_debut`
FROM
`table1` r,
`table2` a,
`table3` p
WHERE
a.`match_id` = m.`match_id` AND
a.`player_id` = p.`player_id` AND
m.`match_void` = '0' AND
(
a.`player_order` BETWEEN '1' AND '11' OR
a.`player_sub_on_for` != '0'
)
GROUP BY p.`player_id`
ORDER BY p.`player_name` ASC
Essentially, as I am making use of the same tables for both queries and only utilising a different WHERE clause, I'm trying to establish if there is a way to 'nest' this.

Answer 1

You may just need conditional aggregation
SELECT p.`player_id`, p.`player_name` AS `name`,
min(case when a.player_order <= '11' OR a.`player_sub` != '0' then r.date else 0 end) `debut`,
SUM(case when a.`player_order` <= '11' OR a.`player_sub` != '0' then 1 else 0 end) AS `apps`,
SUM(case when a.`player_order` <= '11' then 1 else 0 end) AS `starts`,
SUM(a.`player_goals`) AS `goals`
FROM
`table1` r,
left join `table2` a on r.`match_id` = a.`match_id`,
left join `table3` p on a.`player_id` = p.`player_id`
WHERE r.`void` = '0'
GROUP BY p.player_id,a.`player_id`
ORDER BY p.player_id,p.`player_name`;
There seem to be some inconsistencies in your column names ( a.player_sub,a.player_sub_on_for, m.match_void, r.void = '0') so I may not have got this quite right , and group by clause without aggregation is pointless.